REVISED MI DEFINITIONS IMPLICATIONS FOR CLINICAL TRIALS. Maarten L Simoons Thoraxcenter - Erasmus MC Rotterdam - The Netherlands

REVISED MI DEFINITIONS IMPLICATIONS FOR CLINICAL TRIALS Maarten L Simoons Thoraxcenter - Erasmus MC Rotterdam - The Netherlands

TRITON Prasugrel ACS + PCI n = 13,608 moderate / high risk ACS, all PCI p < 0.001 < 0.001 ns ns 0.03 12.1 9.1 9.7 7.4 ClopPras 3.2 3.0 1.0 1.0 1.8 2.4 CV death MI, stroke MI Death Stroke major bleed non-cabg Wiviott NEJM 2007

Cumulative Incidence (%) TRITON Prasugrel ACS + PCI Universal MI Classification 29% 18% -- 24% -- P = 0.0015 0.0015 ns < 0.0001 CLOPIDOGREL PRASUGREL 6.4 4.8 3.4 2.5 0.4 0.3 0.1 0 0 0.1 Spont. Second. SCD PCI CABG

MI IN TRIALS: CEC - INVESTIGATOR P < 0.001 9.7 ns 7.4 4.4 3.3 Clop Pras Serebruany Thromb Haemost 2012

MI IN TRIALS: CEC - INVESTIGATOR P < 0.001 ns P < 0.001 9.7 ns 7.4 4.4 3.3 6.4 5.4 5.9 5.4 Clop Pras Clop Tica Serebruany Thromb Haemost 2012

MI IN TRIALS: CEC - INVESTIGATOR P = 0.009 ns P = 0.04 P < 0.001 12.8 15.7 12.4 14.2 8.3 9.0 10.0 8.0 Plac Abx Plac Epti.

MI IN TRIALS: CEC - INVESTIGATOR Outcome of trial may vary, depending on definitions of primary endpoints, particularly MI The definition should be pre-specified, according to ESC, ACC, AHA, WHF guidelines Similarly pre-specify definitions of bleeding, stroke etc. Both CEC and investigator reported endoints should be reviewed by regulatory agencies: EMA, FDA

UNIVERSAL DEFINITION OF MI ESC, ACC, AHA Eur Heart J 2000 - MI = necrosis caused by ischaemia ESC, ACC, AHA, WHF Eur Heart J 2007 - distinguish different causes (types) of MI ESC, ACC, AHA, WHF Eur Heart J 2012 - more sensitive markers of myocardial necrosis - myocardial necrosis in critically ill (MI or injury) - myocardial necrosis with PCI or CABG

UNIVERSAL DEFINITION OF MI Evidence of myocardial necrosis - Rise and/or fall of markers of necrosis (troponin) with at least one value > 99% URL In clinical setting of myocardial ischaemia - Symptoms - New (presumed new) ST-T changes or LBBB - Development of (new) Q waves - Imaging evidence of loss of viable myocardium or new wall motion abnormality - Intracoronary thrombus by angio / autopsy

MYOCARDIAL NECROSIS Primary myocardial ischaemia (MI type 1) - Plaque rupture / fissure - Intracoronary thrombus Ischaemia by supply / demand imbalance - Tachy-brady-arrhythmia - Aortic dissection, severe Aortic valve disease - Hypertrophic cardiomyopathy - Cardiogenic, hypovolemic, septic shock - Severe anaemia - Hypertension (+/- LVH) - Coronary spasm, embolism, vasculitis - Endothelial dysfunction without significant CAD

MYOCARDIAL NECROSIS Injury not related to myocardial ischaemia - Cardiac contusion, surgery, ablation, pacing, defibrillator shock - Rhabdomyolysis (cardiac involvement) - Myocarditis - Cardiotoxic agents (antracyclines, herceptin) Multifactorial / undetermined myocardial injury - Heart failure, Stress (Takotsubo) cardiomyopathy - Pulmonary embolism, pulmonary hypertension - Sepsis and critical ill patients - Renal failure, stroke, intracranial bleeding - Amyloidosis, sarcoidosis, strenuous exercise

PCI RELATED MI Continuing discussion about relevance of PCI related myocardial injury: - It is often unavoidable, however it is better if such injury could be limited - It can be limited by careful catheter / wire handling and by anti-thrombotic therapy - Even in patients with successful, elective PCI and a normal troponin level before the procedure, troponin elevation (myocardial injury) is frequent - The long-term consequences are uncertain Similarly, in cardiac surgery some injury is often unavoidable

PCI RELATED MI 44 % > 97.5 th percentile 23 % > 3 x 99 th percentile Troponin elevation no impact on prognosis (1 yr death / MI) 0 0.3 1 2 3 4 Cardiac troponin I ng/ml 3200 patients, successful elective PCI De Labriolle Am J Card 2009

Mortality at 6 months follow-up (%) Akkerhuis, Simoons Circulation 2002 PCI related Spontaneous CAPTURE, EPIC, EPILOG IMPACT II, PURSUIT N = 8836 7.4 1.3 2.0 2.3 4.3 PURSUIT N = 5583 4.1 8.6 9.0 14.3 15.5 1 3 5 10 1 3 5 10 CK-MB elevation relative to ULN.

SPONTANEOUS / PCI RELATED MI 1 10 100 13,602 patients, 1118 new or recurrent MI after enrollment, 6 m follow-up TRITON-TIMI 38 Bonaca et al Circulation 2012

SPONTANEOUS / PCI RELATED MI 5467 patients, 212 PCI related MI, 236 spontaneous MI after enrollment, 5 year follow-up. FRISC II, ICTUS, RITA-3 Damman et al Circulation 2012

SPONTANEOUS / PCI RELATED MI Discussion about relevance of PCI related injury: - Elective PCI / ACS; varying follow-up - Inclusion or exclusion of patients with elevated markers of necrosis before PCI - Even if CK or CK-MB was normal before PCI, troponin might have been elevated - Results may depend on method of analysis, some studies were underpowered Consequences of PCI related myocardial injury are less severe than spontaneous MI, which is important for interpretation of trial results and for information to patients. Damman, Wallentin, Fox, Circ 2012

PCI RELATED MI In patients with normal baseline (< 99th %) troponin elevation > 5 x 99th % within 48 hours after the procedure (arbitrary) are indicative of MI, if - Evidence of prolonged ischemia (pain > 20 min) - Ischemic ST changes (>20 min) or new Q waves - Angiographic evidence of flow limiting complications: side branch occlusion, no-reflow, persistent low flow, embolization - Imaging evidence of new loss of viable myocardium or new wall motion abnormality ESC, ACC, AHA, WHF 2012

CABG RELATED MI In patients with normal baseline (< 99th %) troponin elevation > 10 x 99th % within 48 hours after the procedure (arbitrary) are indicative of MI, if - New Q waves or LBBB - Angiographic evidence of graft- or coronary artery occlusion - Imaging evidence of new loss of viable myocardium or new wall motion abnormality ESC, ACC, AHA, WHF 2012

REPORTING MI IN CLINICAL TRIALS Troponin > 99 th % Spontaneous 1 1-3 3-5 5-10 > 10 Total Secondary 2 Death 3 PCI related 4 Stent occl. Restenosis CABG 5

REPORTING MI IN CLINICAL TRIALS Collect blood samples at baseline and 3 6 hours later in patients with possible / suspected MI (ACS). Collect samples before and 3 6 hrs after PCI / CABG Assess cardiac Troponin I / T (If Troponin cannot be measured use CK-MB mass) Interpret using 99 th percentile URL for each laboratory Present the definition of MI to be used in the trial analysis plan and define role of Clinical Event Cie Report MI, and other endpoints. (trial definitions). Also report the full data table (Guidelines) such that alternative definitions can be applied in meta-analyses

REPORTING MI IN CLINICAL TRIALS Troponin > 99 th % Spontaneous 1 1-3 3-5 5-10 > 10 Total Secondary 2 Death 3 PCI related 4 Stent occl. Restenosis CABG 5